src/daComposant/daAlgorithms/LinearLeastSquares.py

   1 # -*- coding: utf-8 -*-
   2 #
   3 # Copyright (C) 2008-2019 EDF R&D
   4 #
   5 # This library is free software; you can redistribute it and/or
   6 # modify it under the terms of the GNU Lesser General Public
   7 # License as published by the Free Software Foundation; either
   8 # version 2.1 of the License.
   9 #
  10 # This library is distributed in the hope that it will be useful,
  11 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13 # Lesser General Public License for more details.
  14 #
  15 # You should have received a copy of the GNU Lesser General Public
  16 # License along with this library; if not, write to the Free Software
  17 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307 USA
  18 #
  19 # See http://www.salome-platform.org/ or email : webmaster.salome@opencascade.com
  20 #
  21 # Author: Jean-Philippe Argaud, jean-philippe.argaud@edf.fr, EDF R&D
  22
  23 import logging
  24 from daCore import BasicObjects
  25 import numpy
  26
  27 # ==============================================================================
  28 class ElementaryAlgorithm(BasicObjects.Algorithm):
  29     def __init__(self):
  30         BasicObjects.Algorithm.__init__(self, "LINEARLEASTSQUARES")
  31         self.defineRequiredParameter(
  32             name     = "StoreInternalVariables",
  33             default  = False,
  34             typecast = bool,
  35             message  = "Stockage des variables internes ou intermédiaires du calcul",
  36             )
  37         self.defineRequiredParameter(
  38             name     = "StoreSupplementaryCalculations",
  39             default  = [],
  40             typecast = tuple,
  41             message  = "Liste de calculs supplémentaires à stocker et/ou effectuer",
  42             listval  = [
  43                 "Analysis",
  44                 "CostFunctionJ",
  45                 "CostFunctionJAtCurrentOptimum",
  46                 "CostFunctionJb",
  47                 "CostFunctionJbAtCurrentOptimum",
  48                 "CostFunctionJo",
  49                 "CostFunctionJoAtCurrentOptimum",
  50                 "CurrentOptimum",
  51                 "CurrentState",
  52                 "OMA",
  53                 "SimulatedObservationAtCurrentOptimum",
  54                 "SimulatedObservationAtCurrentState",
  55                 "SimulatedObservationAtOptimum",
  56                 ]
  57             )
  58         self.requireInputArguments(
  59             mandatory= ("Y", "HO", "R"),
  60             )
  61
  62     def run(self, Xb=None, Y=None, U=None, HO=None, EM=None, CM=None, R=None, B=None, Q=None, Parameters=None):
  63         self._pre_run(Parameters, Xb, Y, R, B, Q)
  64         #
  65         Hm = HO["Tangent"].asMatrix(None)
  66         Hm = Hm.reshape(Y.size,-1) # ADAO & check shape
  67         Ha = HO["Adjoint"].asMatrix(None)
  68         Ha = Ha.reshape(-1,Y.size) # ADAO & check shape
  69         #
  70         RI = R.getI()
  71         #
  72         # Calcul de la matrice de gain et de l'analyse
  73         # --------------------------------------------
  74         K = (Ha * RI * Hm).I * Ha * RI
  75         Xa =  K * Y
  76         self.StoredVariables["Analysis"].store( Xa.A1 )
  77         #
  78         # Calcul de la fonction coût
  79         # --------------------------
  80         if self._parameters["StoreInternalVariables"] or \
  81             self._toStore("CostFunctionJ")  or self._toStore("CostFunctionJAtCurrentOptimum") or \
  82             self._toStore("CostFunctionJb") or self._toStore("CostFunctionJbAtCurrentOptimum") or \
  83             self._toStore("CostFunctionJo") or self._toStore("CostFunctionJoAtCurrentOptimum") or \
  84             self._toStore("OMA") or \
  85             self._toStore("SimulatedObservationAtCurrentOptimum") or \
  86             self._toStore("SimulatedObservationAtCurrentState") or \
  87             self._toStore("SimulatedObservationAtOptimum"):
  88             HXa = Hm * Xa
  89             oma = Y - HXa
  90         if self._parameters["StoreInternalVariables"] or \
  91             self._toStore("CostFunctionJ")  or self._toStore("CostFunctionJAtCurrentOptimum") or \
  92             self._toStore("CostFunctionJb") or self._toStore("CostFunctionJbAtCurrentOptimum") or \
  93             self._toStore("CostFunctionJo") or self._toStore("CostFunctionJoAtCurrentOptimum"):
  94             Jb  = 0.
  95             Jo  = float( 0.5 * oma.T * RI * oma )
  96             J   = Jb + Jo
  97             self.StoredVariables["CostFunctionJb"].store( Jb )
  98             self.StoredVariables["CostFunctionJo"].store( Jo )
  99             self.StoredVariables["CostFunctionJ" ].store( J )
 100             self.StoredVariables["CostFunctionJbAtCurrentOptimum"].store( Jb )
 101             self.StoredVariables["CostFunctionJoAtCurrentOptimum"].store( Jo )
 102             self.StoredVariables["CostFunctionJAtCurrentOptimum" ].store( J )
 103         #
 104         # Calculs et/ou stockages supplémentaires
 105         # ---------------------------------------
 106         if self._parameters["StoreInternalVariables"] or self._toStore("CurrentState"):
 107             self.StoredVariables["CurrentState"].store( numpy.ravel(Xa) )
 108         if self._toStore("CurrentOptimum"):
 109             self.StoredVariables["CurrentOptimum"].store( numpy.ravel(Xa) )
 110         if self._toStore("OMA"):
 111             self.StoredVariables["OMA"].store( numpy.ravel(oma) )
 112         if self._toStore("SimulatedObservationAtBackground"):
 113             self.StoredVariables["SimulatedObservationAtBackground"].store( numpy.ravel(HXb) )
 114         if self._toStore("SimulatedObservationAtCurrentState"):
 115             self.StoredVariables["SimulatedObservationAtCurrentState"].store( numpy.ravel(HXa) )
 116         if self._toStore("SimulatedObservationAtCurrentOptimum"):
 117             self.StoredVariables["SimulatedObservationAtCurrentOptimum"].store( numpy.ravel(HXa) )
 118         if self._toStore("SimulatedObservationAtOptimum"):
 119             self.StoredVariables["SimulatedObservationAtOptimum"].store( numpy.ravel(HXa) )
 120         #
 121         self._post_run(HO)
 122         return 0
 123
 124 # ==============================================================================
 125 if __name__ == "__main__":
 126     print('\n AUTODIAGNOSTIC\n')